Chase Combining HARQ vs Incremental Redundancy HARQ

This article compares Chase Combining HARQ and Incremental Redundancy HARQ techniques, outlining their differences and implementations.

Chase Combining HARQ

In Chase Combining HARQ, the same information and parity bits are retransmitted each time. The retransmission of the same packet helps increase the signal-to-noise ratio (Eb/No) at the receiver due to Maximum Ratio Combining (MRC). This method is sometimes called “repetition coding” because the same coded information is transmitted repeatedly.

At the receiver, previous packets are stored in a buffer. When a retransmitted packet arrives, it’s combined with previously received (potentially erroneous) packets before being sent to the decoder. The most common technique used for combining these packets is MRC.

• Buffer Size: Equal to the number of coded symbols per coded packet.
• Application: WiMAX according to the IEEE 802.16e-2005 OFDMA Physical Layer.

Fig.1, Chase Combining HARQ

Incremental Redundancy HARQ

In contrast to Chase Combining, Incremental Redundancy HARQ generates multiple different sets of code bits for the same information bits in a packet. These different sets are transmitted under varying channel conditions. They consist of different redundancy “flavors” obtained through different puncturing configurations.

This scheme, where additional redundant information bits are transmitted with each re-transmission, and the receiver attempts decoding each time, is known as IR type II HARQ. In type III HARQ, each re-transmitted packet is self-decodable.

• Buffer Size: Equal to the number of coded bits of the total transmitted coded packets.
• Application: HSDPA as per the UMTS R5 standard.

Fig.2, Incremental redundancy HARQ

Chase combining and incremental redundancy techniques are invaluable for engineers working on the physical layer development of advanced wireless technologies like Mobile WiMAX and LTE.